A little bit of venting:I wanted to comment on the situation at the moment with DIGI, and some of the conversation being had over in the Gulden forums. While the discussion is in Dutch, hopefully Google translate isn't giving me the wrong translation. So there are some points I wanted to address.
Op die manier hebben we Digi geimplementeerd, aan de hand van suggesties, ideeen en gevoelens, maar niet aan de hand van bewijzen. (Translated to:In this way we have implemented Digi, according to suggestions, ideas and feelings, but not on the basis of evidence. And it does not work, in fact the risk was even present it would work worse.)
It has proven effective for a large number of altcoins. My team's suggestion for DIGI wasn't based on feelings, it's was based on the fact that DIGI works for a lot of other coins. It's just not working with our block times and the absence of dedicated miners. But I'll come back to the block times in a minute. Anyone else notice that as dedicated mining power reduced over the last week, things got progressively worse? People jumped ship and furthered the situation we're in now. Fact of the matter is we are playing ping-pong with CM. We find a long high block and CM finds the short block. There is no middle ground without dedicated miners. Top that off with my thoughts that we're moving too far in either direction means that we're never landing in a difficulty zone that accommodates our dedicated miners, and we're just struggling more than we need to right now.
Back to block times. DIGI, and other single block retargets, work best with a block time closer to 1 minute. Lets do some math:
Our block retarget = 150 seconds. Let's say the last block came in at 15 seconds. That's 10% of the retarget time. Let's say block retarget = 75 seconds. We have the same last block time of 15 seconds. That's 20% of the retarget time. Regardless of the change limit, you could guess which is getting a bigger proportional change to difficulty.
So with our block times, not only are we hitting extreme retarget/actual block time proportions, but we're using a higher percentage of change.
I'll admit that DIGI is a little wonky right now. However, it is the right algo for the situation. With dedicated mining power, DIGI will work in it's current form. CM will take less coins, and miners will make more profit.
De tijd die GJ in de Digi implementatie heeft gestoken, heeft hij niet in de simulator of custom algoritme kunnen steken. Het lijkt dat het verwarrend blijft voor velen. (Translated to: The time GJ has put into the Digi deployment, he was unable to stabbing in the simulator or custom algorithm.)
The translation is a little confusing, but there's one part that really stands out, even with the broken translation- the DIGI work took away from /GJ's ability to work on a custom algorithm. First thing to note about this is that we still haven't seen any results from the simulator. The last github commit was over a month ago. I still don't see the DIGI code in the simulator. While I would love /GJ's simulator to be the end all be all source of a new or custom algo, we haven't seen anything from it other than false hope. I know that sounds harsh, but it's how I feel. The simulator is still unproven. It will be so until it's tested against a testnet, and the results are reviewed by someone other than it's creator.
Now if any of the translations were just totally off, and I'm reading it the wrong way, please let me know.
Speaking of simulators:That being said, I wanted to share the latest developments out of the Criptoe team. We are still committed to providing any and all support we can for the coin. And while we agree that the simulator is a great idea, we're just GO dumb. So instead of learning GO, we wrote our own simulator. Let me rephrase that... our most recent addition to the team wrote our own simulator. So I would like to share his project with everyone now. I'll let him comment on the details when he gets online and sees this.
The simulator is an Excel spreadsheet running a VBS macro scrypt on the backend. It's relatively easy to tweak the numbers to test out different situations, and there are already plans in the works to try to port it to C++ for easy algo code drop in. We know that there are different paths that one can take to get to a solution, so we forged our own instead of waiting for someone else.
So without further ado, I present the Criptoe Mining Simulator:
Settings page:
Mining results page:
European/NL Criptoe Mining Pool:In addition to our simulator, within the next couple weeks, the above mentioned member will be opening a European extension to the Criptoe pool. It will run separately from Criptoe.Com, under his control, and provide a backup pool to those looking for the reliability and trustworthiness that the Criptoe team strives for. We'll keep people up to date with launch info, so stay tuned.
I'll leave with this- You may love the Criptoe Team or hate us. You may think we're a bunch of old opinionated bastards, 24Kilo especially(love ya, mate), but we're here to help in any way we can. We've been around long enough to see enough failures in crypto to know when you have a winner. NLG is our winner.
Here's to the success of NLG, from now until the zombie robot apocalypse.
-Fuse
Excellent work Mr Fuse, will be interested to see your teams results. I personally appreciate the support you keep giving Guldencoin and I am team Criptoe fan!
Show Posts
Now it's Guldens turn
Works like a charm 
